ti.get_previous_ti method is not optimized

### Under which category would you file this issue?

Airflow Core

### Apache Airflow version

3.3.0

### What happened and how to reproduce it?

When we execute `ti.get_previous_ti(state='success')` airflow collects all task instances from the database and then gets the latest on api side. It's not effective, because the query itself already has sorting by the date

### What you think should happen instead?

We can add limit 1 to the query and prevent api load

### Operating System

_No response_

### Deployment

None

### Apache Airflow Provider(s)

_No response_

### Versions of Apache Airflow Providers

_No response_

### Official Helm Chart version

Not Applicable

### Kubernetes Version

_No response_

### Helm Chart configuration

_No response_

### Docker Image customizations

_No response_

### Anything else?

_No response_

### Are you willing to submit PR?

- [x] Yes I am willing to submit a PR!

### Code of Conduct

- [x] I agree to follow this project's [Code of Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)

Start by locating the Python implementation of `ti.get_previous_ti` and inspect how its database query orders and retrieves task instances. Find the existing tests for this method, then verify that the query returns only the latest matching task instance and that the tests pass without changing behavior.

Written by the indexing model from the issue text.